Learning Automata - A Survey

By K. Narendra, M. Thathachar. 1974

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Deep learning in neural networks: An overview (2014, background): “The following decades brought simple NNs trained by SL (e.g., Narendra & Thathatchar, 1974; Rosenblatt, 1958, 1962; Widrow & Hoff, 1962) and UL (e.g., Grossberg, 1969; Kohonen, 1972; von der Malsburg, 1973; Willshaw & von der Malsburg, 1976), as well as closely related associative memories (e.g.,…”
Evolutionary Dynamics of Multi-Agent Learning: A Survey (2015, background): “Traditional single-agent reinforcement learning algorithms, such as Q-learning (Watkins & Dayan, 1992) and (networks of) learning automata (Narendra & Thathachar, 1974; Wheeler Jr & Narendra, 1986; Vrancx, Verbeeck, & Nowé, 2008b), can be directly applied in this setting.”
A Minimum Relative Entropy Principle for Learning and Acting (2008, background): “Other stochastic action selection approaches are found in Wyatt (1997) who examines exploration strategies for (PO)MDPs, in learning automata (Narendra and Thathachar, 1974) and in probability matching (R.O. Duda, 2001) amongst others.”
Safe Exploration of State and Action Spaces in Reinforcement Learning (2012, methodology): “This is similar to a linear reward-inaction update for learning automata (Narendra & Thathachar, 1974, 1989) in which the sign of the TD error is used as a measure of success.”
A Bayesian Rule for Adaptive Control based on Causal Interventions (2009, background): “Other stochastic action selection approaches are found in exploration strategies for (PO)MDPs [Wyatt, 1997], learning automata [Narendra and Thathachar, 1974] and in probability matching [R. O. Duda, 2001] amongst others.”
Reinforcement Learning in Continuous State and Action Spaces (2012, methodology): “This is similar to a linear reward-inaction update for learning automata (Narendra and Thathachar, 1974, 1989), using the sign of the temporal-difference error as a measure of ‘success’.”
Reinforcement Learning: An Introduction (1998, methodology): “One of the early breakthroughs in reinforcement learning was the development of an o↵-policy TD control algorithm known as Q-learning (Watkins, 1989), defined by Q(St, At) Q(St, At) + ↵ h Rt+1 + max a Q(St+1, a) Q(St, At) i .”
Evolutionary game theory and multi-agent reinforcement learning (2005, background): “Since then there has been a lot of development in the ﬁeld and a number of survey papers and books on this topic have been published ( Narendra & Thathacher , 1974; 1989; Thathacher , 2002).”
When can the two-armed bandit algorithm be trusted? (2004, methodology): “In this field it is known as the Linear Reward-Inaction (LR−I ) scheme (see the survey [10] and the book [11] by Narendra and Thathachar about learning automata theory).”
An Evolutionary Dynamical Analysis of Multi-Agent Learning in Iterated Games (2005, background): “The term LA was for the first time introduced in the work of Narendra and Thathacher in 1974 [25].”
Varieties of learning automata: an overview (2002, background): “in 1974 [1].”